The RFN riboswitch of Bacillus subtilis is a target for the antibiotic roseoflavin produced by Streptomyces davawensis

Transcription

1 [RNA Biology 6:3, ; July/August 2009]; 2009 Landes Bioscience Research Paper The RFN riboswitch of Bacillus subtilis is a target for the antibiotic roseoflavin produced by Streptomyces davawensis Eva Ott, 1 Jürgen Stolz, 2 Martin Lehmann 3 and Matthias Mack 1, * 1 Institut für Technische Mikrobiologie; Hochschule Mannheim; Mannheim, Germany; 2 Zentralinstitut für Ernährungs-und Lebensmittelforschung (ZIEL); Lehrstuhl für Ernährungsphysiologie; Technische Universität München; Freising-Weihenstephan, Germany; 3 Biotechnology R&D; DSM Nutritional Products; Basel, Switzerland Key words: RFN riboswitches, Bacillus subtilis, riboflavin biosynthesis, roseoflavin, Streptomyces davawensis The riboflavin (vitamin B 2 ) biosynthetic genes in Bacillus subtilis are transcribed simultaneously from the riboflavin promoter (P rib ). The 5'-end of the nascent rib-mrna carries a flavin mononucleotide (FMN) binding riboswitch, which regulates gene expression. The antibiotic roseoflavin from Streptomyces davawensis is a naturally occurring riboflavin analog, its mechanism of action is largely unknown. A recombinant B. subtilis strain carrying a copy of P rib -RFN fused to a promoterless lacz reporter gene in the chromosomal amye locus was grown in a minimal medium. Upon addition of roseoflavin to the growth medium the apparent LacZ activity in this strain was not significantly reduced. Similar experiments carried out on recombinant B. subtilis strains oversynthesizing the flavin transporters RibU (B. subtilis) or RibM (S. davawensis) produced still other results. In these strains, roseoflavin (as well as riboflavin) repressed LacZ synthesis indicating that the RFN riboswitch is a target for roseoflavin (or roseoflavin mononucleotide), which may at least in part explain its antibiotic activity. Introduction In many bacteria, the expression of genes crucial to metabolite biosynthesis (or transport) is regulated by riboswitches. 1,2 Typically found in the 5'-untranslated region (5'-UTR) of certain mrnas, riboswitches form a highly selective receptor (aptamer) and bind a specific metabolite. Metabolite binding to the receptor causes premature transcription termination or precludes access to the ribosomal binding site. Accordingly, riboflavin (RFN) riboswitches sense the concentration of the riboflavin derived cofactor flavin mononucleotide (FMN) and regulate riboflavin biosynthesis. In Bacillus subtilis the riboflavin biosynthetic genes ribgbaht are transcribed from the rib-promoter (P rib ) and the corresponding gene products of the operon synthesize riboflavin from GTP *Correspondence to: Matthias Mack; Institut für Technische Mikrobiologie; Hochschule Mannheim; Paul-Wittsack-Str. 10; Mannheim Germany; Tel.: ; Fax: ; m.mack@hs-mannheim.de Submitted: 02/05/09; Revised: 03/05/09; Accepted: 03/05/09 Previously published online as an RNA Biology E-publication: and two molecules of ribulose-5'-phosphate. 3 FMN binding to the RFN riboswitch stabilizes an mrna structure (terminator) that causes transcription termination before the adjacent ribgenes can fully be transcribed. 4,5 Recent works suggest that some antibacterial compounds function by targeting riboswitches. 6-8 Roseoflavin (Fig. 1) (MIC B. subtilis 1.6 μg/ml), the only known natural riboflavin analog with antibiotic function, is synthesized by Streptomyces davawensis. 9,10 Roseoflavin is toxic to Gram-positive but also to Gram-negative bacteria if the compound is able to enter the cell. 11 The riboflavin transporter RibU (YpaA) is responsible for riboflavin and roseoflavin uptake in B. subtilis. 12 A similar function (flavin uptake) was shown for RibM from S. davawensis. 11 Cytoplasmic roseoflavin is converted to the corresponding FMN/ FAD-analogs roseoflavin-5'-monophosphate (RoFMN) (Fig. 1) and roseoflavin adenine dinucleotide (RoFAD) by the bifunctional flavokinase (EC )/FAD synthetase ( ) RibC. 13,14 For D-amino acid oxidase (EC ) from Sus scrofa it was shown that RoFAD is an inactive cofactor. 13 Consequently, the synthesis of inactive cofactors may at least in part account for the observed antibiotic activity of roseoflavin. 15 However, as hypothesized above, roseoflavin (RoFMN or RoFAD) may in addition target RFN riboswitches. Using P rib -RFN-lacZ fusions in B. subtilis we were able to show that this is very likely the case. Results The addition of riboflavin but not roseoflavin affects the activity of the rib-promoter region in a B. subtilis wild-type strain. To test whether riboflavin/roseoflavin affect RFN in vivo a chromosomal P rib -RFN-lacZ transcriptional fusion was constructed (Fig. 2). The fusion was inserted in the amye locus of B. subtilis wild-type cells (B. subtilis strain peo10) and thus present in a single copy within the chromosome. B. subtilis strain peo10 was pregrown, riboflavin was added and the cells were further incubated. P rib -RFN-driven β-galactosidase synthesis was determined in cell-free extracts of B. subtilis strain peo10 by measuring LacZ activity using ONPG (Fig. 3A). A decrease in specific LacZ activity by 61% indicated that a physiological amount of riboflavin (50 μm) in this system led to a significantly reduced amount of full-length mrna. This experiment was performed by others with similar results RNA Biology 2009; Vol. 6 Issue 3

2 Addition of saturating amounts of riboflavin (>250 μm) had an even stronger effect, LacZ activity was reduced by 79% with comparison to the controls (Fig. 3A). FMN is known to be the principal effector for RFN. 5 In contrast to this previous report, the addition of FMN (50 μm) to our cultures did not significantly reduce LacZ activity (data not shown). This was probably due to the fact that FMN was not transported very efficiently. 12 FMN and RoFMN therefore were not employed in the following experiments. To test the in vivo effect of roseoflavin, B. subtilis strain peo10 was pregrown and roseoflavin was added to the growth medium instead of riboflavin. The addition of roseoflavin (50 μm and >250 μm) only slightly reduced the growth rate of the cultures (data not shown) and had no apparent effect on LacZ activity (Fig. 3A). A B. subtilis control strain containing pdg268 only (lacz without a promoter) did not produce any LacZ activity (data not shown). Increased flavin transport resulted in a repressing effect of roseoflavin. The results above suggested that roseoflavin in vivo was not a target for RFN. Subsequent experiments using recombinant B. subtilis strains overproducing flavin transporters were initiated in order to exclude the possibility that roseoflavin was not entering the cytoplasm in sufficient amounts to modulate RFN activity. Riboflavin transport in B. subtilis was analysed in detail earlier. 12 RibU from B. subtilis is related to the riboflavin transporter RibU from Lactococcus lactis 16 and is a member of the bile/arsenite/ riboflavin transporter (BART) superfamily. In contrast, RibM from Streptomyces davawensis is not a member of this family but is related to RibM (PnuX) from Corynebacterium glutamicum. Overproduction of B. subtilis RibU and S. davawensis RibM were shown to allow growth of riboflavin auxotrophic strains of B. subtilis 12 and E. coli, 11 respectively. A RibU overproducing B. subtilis strain was more active in [ 14 C]riboflavin uptake as compared to a wild-type strain. Also, RibU and RibM overproducing strains were more susceptible to roseoflavin. Furthermore, the uptake of [ 14 C]riboflavin via RibU was significantly reduced in the presence of roseoflavin indicating that roseoflavin is indeed a substrate for B. subtilis RibU. 12 The same P rib -RFN-lacZ construct (peo10) as described above was used to transform a B. subtilis strain containing the gene for the flavin transporter ribm from S. davawensis at sacb under control of the constitutive promoter P vegi. In this strain riboflavin Figure 1. The enzymatic conversion of riboflavin (top) into (ribo)flavin-5'-monophosphate (FMN) and of roseoflavin (bottom) into roseoflavin mononucleotide (RoFMN). Figure 2. Design of the lacz fusion experiments. The lacz gene (in pdg268) was fused with the Bacillus subtilis P rib promoter region (-35 and -10) containing the RFN element (via EcoRI/BamHI sites). The lacz fusion is flanked by forward and backward fragments of the B. subtilis amye gene allowing double crossover integration into the amye site of the B. subtilis chromosome with chloramphenicol (CmR) resistance selection. The transcription start (+1) and the ribosomal binding site (RBS) of lacz are shown. (50 μm/250 μm) had a strong effect on repressing P rib -RFN (reduction to 11%/13% of LacZ activity) and also roseoflavin (50 μm/250 μm) reduced LacZ activity to 17%/13% of control levels (Fig. 3B). Growth of the cultures was reduced upon addition of roseoflavin (final OD was reduced to 75% as compared to riboflavin grown cells). The plasmid peo10 was subsequently used to transform a B. subtilis strain, which allows the IPTG-inducible overproduction RNA Biology 277

3 Figure 3. B. subtilis wild type (wt) (A), RibM (B) and RibU (YpaA) (C) overproducing strains, each carrying a copy of P rib -RFN fused to the promoterless lacz reporter gene (including a ribosomal binding site) in the chromosomal amye locus, were grown in Spizizen minimal medium with glucose (0.4%) and casamino acids (0.02%) at 37 C until the early exponential growth phase. Cells were collected by centrifugation and resuspended in fresh medium containing riboflavin (RF) and/or roseoflavin (RoF) with the final concentration (μm) as indicated in the table (s = saturating amounts; >250 μm). Cells then continued to grow for 4.5 hr at 37 C. The specific activity of β-galactosidase (LacZ) was determined with 2-nitrophenyl-β-Dgalactopyranoside (ONPG) as the substrate and expressed as nanomoles of ONPG hydrolyzed per minute per μg of total protein (mu/μg). The data represent mean values from three independent experiments with the indicated standard deviation. For the RibU (YpaA) (C) overproducing strain IPTG (1 mm) was added to induce ribu expression. of the well characterized flavin transporter RibU (YpaA) from B. subtilis. In this strain in the presence of IPTG (1 mm), riboflavin (50 μm) had a strong effect on repressing P rib -RFN (reduction to 19% of LacZ activity) and also roseoflavin (50 μm) reduced LacZ activity to 32% of control levels (Fig. 3C). Growth as well was reduced upon addition of roseoflavin (final OD was reduced to 80% as compared to riboflavin grown cells). As expected, in the absence of IPTG only riboflavin was active in repressing P rib -RFN (Fig. 3C). Discussion S. davawensis synthesizes roseoflavin, which exhibits antibiotic activity against Gram-positive bacteria. According to our recent work roseoflavin is actively taken up by B. subtilis cells and quickly is converted to RoFMN by RibC Consequently, the addition of roseoflavin to a B. subtilis culture leads to accumulation of RoFMN within the cell. We now found in our experiments using P rib -RFN-lacZ fusions in B. subtilis that the addition of roseoflavin significantly reduced LacZ activity. In light of previous data we conclude that RoFMN (and not roseoflavin) binding to RFN was responsible for inducing premature termination of transcription from P rib -RFN-lacZ as was reported earlier for FMN. 5 The above described observation was made only in recombinant B. subtilis strains with enhanced flavin uptake mediated either by RibU from B. subtilis or RibM from S. davawensis. In a B. subtilis strain without enhanced flavin transport the addition of roseoflavin did not reduce LacZ activity. In the latter case RibU, the endogenous flavin transporter of B. subtilis, probably was not present in sufficient quantity to allow accumulation of repressing amounts of roseoflavin (RoFMN). Furthermore, it was shown earlier that riboflavin was a better substrate for RibU as compared to roseoflavin, 12 which explains why riboflavin did reduce LacZ activity in cells without enhanced flavin transport containing the P rib -RFN-lacZ fusion. In a natural setting, soil microorganisms usually are limited with respect to nutrients. Under these conditions vitamin transporters may be synthesized in larger amounts (as compared to laboratory conditions) to allow uptake of incidentally available growth factors. In light of this, it is plausible that a toxic analog like roseoflavin is able to exhibit its antibiotic activity and help S. davawensis to compete with other soil organisms. The strategy to use an antibiotic structurally related to the vitamin riboflavin, which readily is taken up by competitors, is thereby very efficient ( Trojan horse ). The most recent report on riboswitches 17 shows the three dimensional structure of the RFN riboswitch from the Gramnegative Fusobacterium nucleatum in the presence of riboflavin and also roseoflavin. The structural data revealed, that the roseoflavin-bound structure adopts a conformation similar to the riboflavin-bound structure with only minor additional spatial adjustments to accommodate the dimethylamino group of roseoflavin (Fig. 1). The phosphorylated derivatives of the flavins (RoFMN and FMN) probably behave in a similar way. Thus, the antibiotic activity of roseoflavin may at least in part be due to RoFMN blocking full-length transcription of the rib-genes and producing riboflavin deficient cells. In addition to blocking RFN, RoFMN and RoFAD probably are not active as cofactors for many (if not for all) flavoenzymes. 18 Since roseoflavin also reduces growth of animals (not employing RFN riboswitches), 19 the observed RFN activity cannot be the only explanation for roseoflavin toxicity. 278 RNA Biology 2009; Vol. 6 Issue 3

4 Some roseoflavin resistant strains are deregulated with respect to riboflavin biosynthesis and overproduce riboflavin. 20 These strains carry mutations in either ribc or within the RFN riboswitch: The riboflavin overproducing phenotype in RibC deficient strains can be explained by the reduced synthesis of FMN. 14 Mutations in RFN may prevent FMN from binding to RFN and producing the terminator structure. Other roseoflavin resistant strains interestingly do not oversynthesize riboflavin and probably show mutations in RFN that prevent RoFMN from binding but not FMN. RFN riboswitches in the 5'-UTR of riboflavin biosynthetic genes were detected in Streptomyces avermitilis, 21 Streptomyces coelicolor 22 and S. davawensis. 11 These riboswitches are highly similar but not identical. In contrast to S. davawensis, S. avermitilis and S. coelicolor are roseoflavin sensitive. Roseoflavin resistance of S. davawensis may (in part) be mediated by an RFN riboswitch, which does not bind RoFMN. Materials and Methods Bacterial strains, plasmids and growth conditions. Escherichia coli DH5α 23 was used as a host for gene cloning experiments and was aerobically grown at 37 C on Luria-Bertani (LB) medium. Bacillus subtilis 168 (trpc2) 24 is a wild-type strain with respect to riboflavin biosynthesis and uptake and was used as a host for the lacz fusion experiments. The B. subtilis strain overproducing RibU was generated using the episomal plasmid pdg148-stu replicating in Bacillus from the pub110 origin. 12 The expression of ribu was stimulated by adding 1 mm isopropyl-β-d-thiogalactopyranoside (IPTG). The B. subtilis strain constitutively overproducing RibM from Streptomyces davawensis 13 was generated using the integrative (@sacb) plasmid pxi B. subtilis was aerobically cultivated at 37 C in LB medium, or Spizizen minimal medium supplemented with glucose (0.4%), casamino acids (0.02%) and trypthophan (0.05 mg/ml). When required 5 μg ml -1 chloramphenicol, 5 μg ml -1 kanamycin or 100 μg ml -1 ampicillin were added to the media. Construction and analysis of chromosomal transcriptional P rib -RFN-lacZ fusions. For generation of P rib -RFN-lacZ fusions pdg268, 26 was employed. This plasmid has a promoterless lacz gene and a chloramphenicol (Cm) resistance gene (chloramphenicol acetyl transferase) flanked by forward and backward fragments of the B. subtilis amye gene allowing lacz fusion integration into the amye site of B. subtilis chromosome. The rib-promoter region including the RFN riboswitch from B. subtilis was amplified by PCR using the modifying oligonucleotides RFN_fw (5'-GCT GAA TTC ATC ACC TTT CGG ATC GAA GG-3') and RFN_rv (5'-GCA GGA TCC GTT TCC CTC CCC TCT TTT G-3') (restriction endonuclease sites used for cloning are underlined). As a template for PCR amplification chromosomal DNA of B. subtilis 168 was used. The resulting PCR product was treated with EcoRI and BamHI and ligated to EcoRI and BamHI digested pdg268. E. coli DH5α was transformed with the ligation product. For purification of the corresponding plasmid peo10 a standard protocol 23 was employed. The construct peo10 (1 μg) was linearized using XhoI and used to transform B. subtilis according to a standard procedure. 27 The resulting strain is referred to as B. subtilis peo10 throughout the text. For lacz fusion experiments B. subtilis was cultivated in minimal medium to an OD 600 of Subsequently, flavins (riboflavin or roseoflavin) were added and the cultures were grown for another 4.5 h. The B. subtilis cells were collected by centrifugation, washed in PM (10 mm NaH 2 PO 4, 90 mm Na 2 HPO4, 1 mm MgSO 4, ph 7.8), resuspended in 0.05 volumes PM and disrupted using a vibratory tube mill at maximum speed in the presence of glass beads (0.3 mm in diameter). The specific activity of β-galactosidase was determined with 2-nitrophenyl-β-D-galactopyranoside (ONPG) as substrate and expressed as nanomoles of ONPG hydrolyzed per minute per μg of total protein (mu/μg). Protein was determined according to the method of Bradford. 28 Acknowledgements This work was supported by the Landesstiftung Baden- Württemberg. We thank S. Düsterhus for providing us with pdg268. This work is dedicated to Uta Müller. References 1. Nudler E, Mironov AS. The riboswitch control of bacterial metabolism. Trends Biochem Sci 2004; 29: Winkler WC, Breaker RR. Regulation of bacterial gene expression by riboswitches. Annu Rev Microbiol 2005; 59: Perkins J, Pero J. Biosynthesis of riboflavin, biotin, folic acid and cobalamin. In: Sonenshein A, Hoch J, Losick R, eds. Bacillus subtilis and Its Closest Relatives: from Genes to Cells. Washington DC: ASM Press 2002; Winkler WC, Cohen-Chalamish S, Breaker RR. An mrna structure that controls gene expression by binding FMN. Proc Natl Acad Sci USA 2002; 99: Mironov AS, Gusarov I, Rafikov R, Lopez LE, Shatalin K, Kreneva RA, et al. Sensing small molecules by nascent RNA: a mechanism to control transcription in bacteria. Cell 2002; 111: Blount KF, Breaker RR. Riboswitches as antibacterial drug targets. Nat Biotechnol 2006; 24: Blount KF, Wang JX, Lim J, Sudarsan N, Breaker RR. Antibacterial lysine analogs that target lysine riboswitches. Nat Chem Biol 2007; 3: Sudarsan N, Cohen-Chalamish S, Nakamura S, Emilsson GM, Breaker RR. Thiamine pyrophosphate riboswitches are targets for the antimicrobial compound pyrithiamine. Chem Biol 2005; 12: Otani S, Takatsu M, Nakano M, Kasai S, Miura R. Letter: Roseoflavin, a new antimicrobial pigment from Streptomyces. J Antibiot (Tokyo) 1974; 27: Mack M, Grill S. Riboflavin analogs and inhibitors of riboflavin biosynthesis. Appl Microbiol Biotechnol 2006; 71: Grill S, Yamaguchi H, Wagner H, Zwahlen L, Kusch U, Mack M. Identification and characterization of two Streptomyces davawensis riboflavin biosynthesis gene clusters. Arch Microbiol 2007; 188: Vogl C, Grill S, Schilling O, Stulke J, Mack M, Stolz J. Characterization of riboflavin (vitamin B2) transport proteins from Bacillus subtilis and Corynebacterium glutamicum. J Bacteriol 2007; 189: Grill S, Busenbender S, Pfeiffer M, Kohler U, Mack M. The bifunctional flavokinase/ flavin adenine dinucleotide synthetase from Streptomyces davawensis produces inactive flavin cofactors and is not involved in resistance to the antibiotic roseoflavin. J Bacteriol 2008; 190: Mack M, van Loon AP, Hohmann HP. Regulation of riboflavin biosynthesis in Bacillus subtilis is affected by the activity of the flavokinase/flavin adenine dinucleotide synthetase encoded by ribc. J Bacteriol 1998; 180: Otani S, Note Y, Nishina Y, Matsumura Y. Interactions between roseoflavin and flavoproteins. In: Yagi K, Yamano T, eds. Flavins and Flavoproteins. Tokyo: Japan Scientific Societies Press 1980; Burgess CM, Slotboom DJ, Geertsma ER, Duurkens RH, Poolman B, van Sinderen D. The riboflavin transporter RibU in Lactococcus lactis: molecular characterization of gene expression and the transport mechanism. J Bacteriol 2006; 188: Serganov A, Huang L, Patel DJ. Coenzyme recognition and gene regulation by a flavin mononucleotide riboswitch. Nature 2009; In press. 18. Hasford J, Rizzo C. Linear free energy substituent effect on flavin redox chemistry. J Am Chem Soc 1998; 120: Otani S, Matsui K, Kasai S. Chemistry and biochemistry of 8-aminoflavins. Osaka City Med J 1997; 43: RNA Biology 279

5 20. Matsui K, Wang H, Hirota T, Matsukawa H, Kasai S, Shinagawa K, et al. Riboflavin production by roseoflavin-resistant strains of some bacteria. Agric Biol Chem 1982; 46: Ikeda H, Ishikawa J, Hanamoto A, Shinose M, Kikuchi H, Shiba T, et al. Complete genome sequence and comparative analysis of the industrial microorganism Streptomyces avermitilis. Nat Biotechnol 2003; 21: Vitreschak AG, Rodionov DA, Mironov AA, Gelfand MS. Regulation of riboflavin biosynthesis and transport genes in bacteria by transcriptional and translational attenuation. Nucleic Acids Res 2002; 30: Sambrook J, Fritsch E, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Saito H, Shibata T, Ando T. Mapping of genes determining nonpermissiveness and hostspecific restriction to bacteriophages in Bacillus subtilis Marburg. Mol Gen Genet 1979; 170: Hümbelin M, Griesser V, Keller T, Schurter W, Haiker M, Hohmann HP, et al. GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone 4-phosphate synthase are rate-limiting enzymes in riboflavin synthesis of an industrial Bacillus subtilis strain used for riboflavin production. J Ind Microbiol Biotechnol 1999; 22: Antoniewski C, Savelli B, Stragier P. The spoiij gene, which regulates early developmental steps in Bacillus subtilis, belongs to a class of environmentally responsive genes. J Bacteriol 1990; 172: Kunst F, Rapoport G. Salt stress is an environmental signal affecting degradative enzyme synthesis in Bacillus subtilis. J Bacteriol 1995; 177: Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: RNA Biology 2009; Vol. 6 Issue 3